Introduction: From a larger study examining policy and program information on how Canadian provinces integrate care services, this study aimed to create “priority lists” of 10–15 services that are “absolutely needed” for care integration.

Methodology: A diverse group of over 50 Canadian stakeholders participated in virtual consensus-building using the nominal group technique and a modified e-Delphi method to identify services that focused on two different groups: children and youth with high functional health needs and older adults in functional decline.

Results: Three lists – containing services, processes and infrastructure elements – emerged: one per tracer condition group and a consolidated list. The latter identified the following five services as top priority for primary care integration: mental health and addictions services; home care; transition between urgent-emergency-acute care; medication reconciliation in community pharmacies; and respite care. No single social service was a clear priority, but those that mitigate material deprivation emerged within the top 10.

Discussion: This humble pan-Canadian study shows that priority services in health and social services are neither well integrated nor connected to primary care. It also suggests that effective policy strategizing for primary care integration for those with complex care needs may require thinking beyond the logic of services – given their siloed organization.

Introduction

People with complex care needs benefit greatly from the integration of their care across health and social service sectors (Bayliss et al. 2015; Breneol et al. 2022; Jones et al. 2020; McCormack et al. 2008; Miller et al. 2009). We defined service integration policies as “the governance, funding, strategic, organizational and programmatic directives that aim to facilitate coordination and collaboration among organizations and service providers whose complementary actions work together to achieve desired health outcomes for patients with complex care needs” (Haggerty et al. 2023: 12; Villeneuve 2017); see Appendix 2, available online here - (version française ici). However, it faces complex structural and infrastructural challenges, notably when care services are organized in silos as in Canada. Patients with complex care needs also often depend on the interprofessional collaboration of those with a multiplicity of expertise (Hutchison et al. 2001). Unfortunately, these complexities have not been properly addressed by healthcare managers and policy makers (Atun and Menabde 2008). Identifying which health, social and community services are critical starting points of integration to primary care for patients with complex care needs is a useful first step in the assessment and development of multiservice integration programs, interventions and policy priorities.

This article is an attempt to make that first step. Here, we describe the methods and results of a consensus development process that took place in 2017, designed to identify, at a minimum, the most important health and social services to integrate with primary care to ensure the provision of comprehensive care for patients with complex care needs. The method chosen was applied to two different groups of such patients: children and youth (0 to 25 years) with high functional health needs and community-dwelling adults (≥ 65 years) experiencing functional decline¹. These groups were chosen to illustrate how, through this method, a consensus among a diverse group of stakeholders from across Canada could be developed and used to better understand the challenges facing service integration in this country. The study was conducted with 50 stakeholders who were, at the time, members of the Canadian Primary Care Research Network (CPCRN) – an interdisciplinary and intersectoral network involving a diversity of stakeholders across sectors, including citizens and service users, who collaborate in generating solution-focused innovations in primary healthcare delivery. The CPCRN's mission is to develop and support “provincial and territorial practice-based research and learning networks” (PBRLNs) (CPCRN n.d.). Patients, clinicians, decision makers and researchers from the CPCRN represented all 10 Canadian provinces.

Methods

Study design

To develop consensus on health and social service categories to be integrated with primary care, we used a two-phase virtual process. First, we used the nominal group technique to generate a list of candidate services for each tracer group (Delbecq and Van de Ven 1970; Gallagher et al. 1993; McMillan et al. 2014); then, we prioritized this list using a modified e-Delphi method (Crisp et al. 1997; Linstone and Turoff 1975). The Delphi method is a widely used and accepted technique well-suited for consensus building. We used a modified Delphi method (Miller and Crabtree 1999) with the goal of using an iterative process with community members to narrow topics to a practical priority list for which there is general consensus (not unanimous agreement) (Addison 1999). This two-phase process was completed in parallel for each tracer group and was integrated at the end of each of the two stages (the end of the nominal group process: to agree on service categorization, labels and definitions; the end of the consensus process: to compare the resulting lists of prioritized services, commonalities and differences). This entire process took place over a compressed time window of 12 weeks (between June and August 2017) to ensure maximal participant engagement. The study received research ethics approval at St. Mary's Research Centre in Montréal. All group members provided participant consent.

Participants

All participants (n = 40), representing Canada's 10 provinces, were invited to participate on the basis of their expertise in or experience of service integration in their respective provinces. Each participant had expertise in research, clinical practice (n = 28) or policy making (n = 9) or they had experiential knowledge as patients or caregivers (n = 3). For this study, the study partners were asked to participate in the tracer condition group (children and youth or older adults) in which they had expertise/experience. For each of the two groups, the entire process was run in parallel with a different set of stakeholders and research teams. All materials were prepared in both French and English, and participants were able to select their language of preference. All communication with participants was electronic and asynchronous via e-mail, an online discussion forum and online surveys.

Data collection and analysis

Nominal Group Process

During a shared forum that lasted 10 days, participants were asked to name the health and social care services that, according to their area of expertise/experience, should be connected to primary care to meet the health and social needs of patients with each tracer condition. To prompt participants to think widely and inclusively about these needs, they received two patient vignettes per tracer condition that described, in a short narrative format, the lived experience of one member of this patient group. The vignettes did not aim for exhaustiveness of experiences or to already name the services a person may need. Rather, they encouraged our participants to think about “people,” and needs in relation to the person's experiences and aspirations (from the point of view of persons, rather than fragmented needs). These vignettes were developed with the study's patient partners and validated by our other study partners with lived experience (either as patients, caregivers or health professionals). Participants were also asked to think beyond the vignettes with their own personal or professional lived experiences using a person- and family-centred perspective. The goal here was to be as inclusive as possible.

Analysis of the Nominal Group Process

The data for each tracer condition were analyzed in parallel by two different research teams composed of two senior researchers, two to three research assistants (graduate studies) and two to three non-research partners (patients, caregivers or health professionals) in order to extract all nominated services and consolidate those that were similar. This team completed an immersion-crystallization phase in which they reviewed the data to identify themes for organization (Borkan 1999), develop a specific coding structure, code text segments according to this structure (Huberman and Miles 2002; Miller and Crabtree 1999), identify patterns and relationships across codes and themes, connect themes and patterns to existing knowledge (Miller and Crabtree 1999) and then corroborate/legitimate the data by seeking out additional data to confirm or refute insights from the initial analysis. Then, using the responses of the participants (n = 23) and relevant information produced by international and national health organizational bodies (World Health Organization and Health Canada websites), all the information provided by participants was analyzed, collapsed into 31 distinct service category labels and labels as well as definitions and examples were provided. This list was then used for the second phase (see Appendix 2, available online here - version française ici - and the “Results” section below).

Modified Delphi

We used Qualtrics, an online service platform, to complete four rounds of an e-Delphi process and prioritize services. This was done by rating and then ranking the service categories.

In the first round, participants (n = 36) were asked to indicate, on a scale of one (not at all) to five (completely), the extent to which they agreed with the service categories (label/name, definition and example). If they did not agree, they were asked to propose modifications, which were integrated by the research teams. The list was considered “accepted” when it was approved by all participants.

In rounds two and three, participants (n = 31, n = 30) were asked to rate the importance of integrating or connecting each service category to primary care by using a RAND scale of one to nine based upon (1) the frequency at which the service was needed by each tracer group and (2) the time-sensitivity required for the connection to avoid negative consequences. Services with mean values between one and three were categorized as “low priority services” (rarely needed or with flexible timeliness). Services scoring between seven and nine were considered “high priority” – either the frequency or time sensitivity was judged critical. Scores were averaged over all participants.

The final round presented participants (n= 40) with the resulting top 15 prioritized services for each tracer condition and asked them to validate the following:

1. the top five services for each tracer condition as a comprehensive and appropriate top-priority service list that would subsequently be used to inform the next studies within the larger research program;
2. the top 15 services per tracer condition;
3. a consolidated list integrating both tracer conditions using the top 10 priority services that emerged from both virtual processes;
4. the comprehensive list of service categories that included validated labels and definitions by all participants; this list emerged from the nominal group process.

With regard to the first three lists, they were asked “Can you live with this list of priorities?” Once finalized and approved, a final report and the lists were shared with all research participants and stakeholders in the study (this took place in the fall of 2017).

Results

The nominal group process yielded 100 services² for children and youth and 31 for older adults. Taken as a distinct service category, we found that no single social service emerged as frequent or time sensitive for either patient group (e.g., educational services, financial services, legal services, etc.). In order to reconcile the need to ensure full consideration of social services with our study design of generating a list of top 10 priority services, social services were grouped into the need they address – that is, material or social deprivation (see Appendix 2, available online here  - version française ici).

Table 1 presents the top 15 services for children and youth. In the service identification and prioritization exercise, telehealth and information-sharing systems were the two highest-priority services rated by children and youth stakeholders. Yet they are not services, per se: health information systems refer to infrastructure while telehealth is a virtual care modality. These were collapsed into a single category of “information-technology enabled care” to render visible the high importance placed on this mechanism for integrated care for children and youth with high functional health needs.

Table 2 presents the top 15 services for older adults, including the ranking of the top 10 services. Eight of the top 10 services are clearly in the traditional health (or medical) domain. Regrouping social services into services addressing material and social deprivation allowed the importance of social services to emerge for this group. The most important material deprivation services are the same for older adults as for youth.

Five services were ranked in the top 10 for both children and youth and older adults: (1) mental health and addictions services, (2) home care, (3) timely transition between urgent, emergency and acute care, (4) medication reconciliation in community pharmacy programs and (5) respite care. All but one (mental health and addictions services) are in the traditional health domain, but they represent a broad range of needs.

Finally, we created a consolidated list of the top 10 services for both groups by averaging the mean ranking across groups and then re-ranking them (e.g., mental health and addictions: second for children and youth and fifth for older adults = 2 + 5 = 7 / 2 = 3.5 overall). Mental health and addictions services emerged as the top priority across both groups by a significant margin (Table 3). However, the most important services to be connected to primary care are in the traditional health domain.

Discussion

Before we begin our discussion of points of relevance to policy making, we have three remarks to formulate.

Firstly, this study was conducted before the COVID-19 pandemic. This crisis has illustrated the need for service integration and provided examples of how this can be done through increased use of information technology services and telehealth initiatives. However, such integration has not been optimized and in some cases, the move to new methods to deliver healthcare has exacerbated problems integration seeks to address. The method employed and the consensus reached in this study remains relevant to identifying the top priority services that should absolutely be integrated or systemically connected with primary care in order to respond to the needs of patients requiring complex care and to support their flourishing.

Secondly, an important discussion that animated this entire study concerns the significant differences in the lexicon and terminology used across Canadian provinces. This proved a conceptual challenge, at first, because even the definition and understanding of “primary care” differed (see Haggerty et al. 2023). This challenge highlights the importance of coming to an informed agreement on the definitions and labels used to identify each service category. This, in our view, suggests that cross-provincial and territorial conversations are needed to achieve care integration, notably given the mobility of Canadians who often need to travel or relocate to other provinces.

Finally, during the nominal group process, participants often identified elements for integration that were not, in fact, services but rather processes, types of professional titles, programs and even elements of infrastructure. In our view, this suggests that we should not think about the integration of comprehensive care solely from the point of view of services. As Kreindler et al. (2022) remark, services are organized in overly siloed ways, and starting from their purview may hinder a truly imaginative process to rethink comprehensive care. Working from patients' needs, rather than from existing services, may be a more fruitful approach to developing care models and orienting policy.

Policy recommendations

Three main findings from this study can, in our view, help inform change in policy and practice. First, five of the top 10 priorities were the same for both groups, and these are all medical services (under the purview of ministries of health). Second, mental health and addictions services emerged as a priority both for children and youth and for older adults. Third, the high priorities placed on home and respite care entail the need for caregivers located outside traditional health service sites (e.g., hospitals, clinics, etc.).

Despite our explicit focus on prioritizing health and social services, our consolidated list of services shows that five of the top 10 priorities are health services under the purview of health ministries. In our view, the continuing predominance of traditional health services illustrates two things. This suggests a continued dominance of downstream rather than upstream thinking by policy makers and service organizations. The effect is that the organization and prioritization of services are done from the purview of managing illness (or health consequences) rather than preventing their emergence. Thinking about integration from a life perspective, rather than from the point of view of short-term acute care for persons with complex needs, may be a more promising and innovative approach to comprehensive care that includes consideration for health promotion and prevention.

The continued dominance of traditional health services also attests to the fact that regardless of recent primary healthcare reforms and policy discourse concerning services and care, the Canadian context of services continues to be organized in silos while primary care is expected to play the role of integration. Yet given this siloing, integration remains an individual, ad hoc and piecemeal task at best, one that falls on the shoulders of primary care providers. The premise that it is the role of primary care to be the nexus where prioritized services are connected (the “medical home”), which is supported by the literature (Cooley 2004; Perrin et al. 2007; Valentijn et al. 2013), was questioned by study participants, specifically in terms of how viable this approach may be in the long term. This concern, which has been highlighted by the ongoing impact of the COVID-19 pandemic, has been echoed in the literature (e.g., Hutchison et al. 2011; Kearon and Risdon 2020; Snelgrove et al. 2022a, 2022b).

The consolidated priority list illustrates the high priority given to mental health and addictions services and highlights important considerations for policy makers. Increased attention has been drawn to mental health and addictions issues in the general population during the COVID-19 pandemic in tandem with the opioid crisis, which predated the pandemic (Gadermann et al. 2021; Regehr et al. 2021; Schmitz et al. 2020; Shatla et al. 2020; Xiong et al. 2020). The stressors experienced by the general population are exacerbated when combined with the challenges that patients with high functional health needs and their caregivers face on a day-to-day basis as they navigate complex health concerns through complex and fragmented health and social systems. We also know from practice and through the scientific literature that unaddressed or unresolved mental health needs may have dire long-lasting effects on patients' or caregivers' health, negatively affect their access to proper healthcare and/or force them to avail themselves of services that would not have otherwise been needed, thus placing an additional burden on the healthcare system (Kieling et al. 2011; Mesidor et al. 2011; Murphy et al. 2017; Patton et al. 2012). Across Canada, strategies to address burgeoning mental health and addictions issues among youth reflect an awareness that an integrated, contextualized health and social services response is required. Such integrated service models provide examples to guide future action (Abidi 2017; Fusar-Poli 2019; Halsall et al. 2019; Mathias et al. 2022; Settipani et al. 2019).

Finally, high rankings for home care and respite care in both groups reflect the dependence of those who live with high functional health needs, regardless of age, on the work of caregivers, be they personal attendants, family members or others (e.g., friends, colleagues). It is erroneous to assume that such support is readily available through existing family and friend networks. Many are without significant others or a personal network, and for those who have relationships, it is not a given that these people are both available and capable to provide the physical and/or psychological support required for adequate care. When the burden to provide such care falls upon understaffed professional care providers or friends and family members – as is more often the case for more marginalized populations who lack access to such services – it can lead to negative health and economic consequences for them, which could have been prevented if they were better supported (Berglund et al. 2015; Berry et al. 2017; Kadowaki et al. 2015; Kok et al. 2015; Kokorelias et al. 2020; Stall 2019; Vandepitte et al. 2016). It should be noted that this burden has only increased in light of the COVID-19 pandemic. In light of this, policy makers should reflect on the demand that is made on patients to equip themselves with a private and personal support network and prioritize the formalization and organization of the support and services provided to all types of caregivers to ensure that they can perform what is demanded of them.

Limitations

There are a number of limitations to this study and on the impact and applicability of our findings. The consensus process was conducted over a short period to maximize engagement and maintain momentum, but this may have entailed some sacrifice of representation from relevant stakeholders. For example, we were unsuccessful in including stakeholders from the Territories (Yukon, Northwest Territories and Nunavut). Additionally, while all provinces were represented, participants from Ontario and Saskatchewan (n = 4) joined the study toward the end of the nominal group process and contributed to the approval of the final lists but not their initial creation.

Secondly, the list was used as the backbone for other components of our research program and continued to be scrutinized and clarified over the life of the research. We did not censor nominated services or activities, and consequently not all “services” can really be classified as services that can be the object of inter-organizational operating policies to integrate care. However, we are confident that the results reflect the importance of different aspects of integrated care for our stakeholder groups. The fact that few individual social services emerged as priority services may show that our initial patient vignettes did not reflect the social determinants of health strongly and may have induced our participants to think more about health issues than about functional capacity more broadly. In the process of consolidating services under a common broad category to render visible social services as a priority for integration with primary healthcare, the importance of specific services may have been diluted and many compromise specificity for policy action.

Furthermore, it remains unclear, with our data, whether the priority given to mental health and addictions services reflects the prevalence of mental health and addictions issues among the patient groups or mental health and addictions issues that develop among patients, families and caregivers due to inherent challenges in providing care to people with high functional health needs or both. Whatever the origin, it is impossible to dismiss the enormous societal ramifications of limited access to much-needed mental health and addictions services and the importance of their integration with primary care and other health and social services (Kieling et al. 2011; Rehm et al. 2006).

Conclusion

The results of this pan-Canadian prioritization of health and social services for integration underscores the need for cross-jurisdictional, national and interprovincial dialogue and decision making to achieve health accords on fundamental health and social infrastructure to ensure equitable access to services for all Canadians. The 2004 Health Accords on home care were a good start (Health Canada 2004), but as demonstrated during the COVID-19 pandemic, huge policy gaps left many of the most vulnerable of our population in extremely precarious, if not life-threatening, situations. We believe a similar process of developing priority lists should include policy makers, professionals, patients and their families from a wider array of sectors. This requires a deeper understanding of patients' lived experiences, their service trajectories and the barriers and facilitators they experience in meeting their day-to-day needs. While innovative approaches to these issues abound, the political, organizational and professional will to transform our approach to health prevention, management and service delivery is required to move beyond firmly entrenched institutional boundaries that stand in the way of achieving health for all.

Correspondence may be directed to Émilie Dionne by e-mail at emilie.dionne3.ciussscn@ssss.gouv.qc.ca.

Vers une intégration complète des soins au Canada : résultats de la méthode de Delphes selon des chercheurs, des cliniciens, des patients et des décideurs

Résumé

Introduction : À partir d'une étude plus vaste qui examine l'information sur les politiques et les programmes visant l'intégration des services de soins dans les provinces canadiennes, cette étude avait pour objet de créer des « listes prioritaires » pour les 10 à 15 services qui sont « absolument nécessaires » pour l'intégration des soins.

Méthode : Un groupe diversifié de plus de 50 intervenants canadiens a participé à l'établissement d'un consensus virtuel, à l'aide de la technique du groupe nominal et d'une méthode de Delphes en ligne modifiée, afin de déterminer quels sont les services prioritaires pour deux groupes différents : d'une part, les enfants et les jeunes ayant de forts besoins en matière de soins fonctionnels et, d'autre part, les personnes âgées en déclin fonctionnel.

Résultats : Trois listes (contenant des services, des processus et des éléments d'infrastructure) se sont dégagées : une pour chacun des groupes de conditions de traçage ainsi qu'une liste consolidée. Cette dernière a permis d'identifier les cinq services suivants comme prioritaires pour l'intégration des soins primaires : les services de santé mentale et de lutte contre les dépendances, les soins à domicile, la transition entre les soins d'urgence et les soins de courte durée, le bilan comparatif des médicaments dans les pharmacies communautaires et, pour terminer, les soins de relève. Aucun service social n'était clairement prioritaire, mais ceux qui atténuent la défavorisation matérielle ont émergé parmi les 10 premiers.

Discussion : Cette humble étude pancanadienne montre que les services prioritaires en santé et en services sociaux ne sont ni bien intégrés ni reliés aux soins primaires. Elle suggère également qu'une politique stratégique efficace pour l'intégration des soins primaires pour les personnes ayant des besoins complexes pourrait nécessiter de penser au-delà de la logique des services, compte tenu de leur organisation cloisonnée.

Comments

Be the first to comment on this!